Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

A single-cell and single-nucleus RNA-Seq toolbox for fresh and frozen human tumors.

Author(s): Slyper M, Porter CBM, Ashenberg O, Waldman J, Drokhlyansky E, Wakiro I, Smillie C, Smith-Rosario G, Wu J, Dionne D, Vigneau S, Jané-Valbuena J, Tickle TL, Napolitano S, Su MJ, Patel AG, Karlstrom A, Gritsch S, Nomura M, Waghray A, Gohil SH, Tsankov AM, Jerby-Arnon L, Cohen O, Klughammer J, Rosen Y, Gould J, Nguyen L, Hofree M, Tramontozzi PJ, Li B, Wu CJ, Izar B, Haq R, Hodi FS, Yoon CH, Hata AN, Baker SJ, Suvà ML, Bueno R, Stover EH, Clay MR, Dyer MA, Collins NB, Matulonis UA, Wagle N, Johnson BE, Rotem A, Rozenblatt-Rosen O, Regev A

Publication: Nat Med, 2020, Vol. 26, Page 792-802

PubMed ID: 32405060 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   The purpose of this study was to investigate potential effects associated with different dissociation methods on single cell RNA sequencing (scRNA-Seq) of fresh tumor specimens. Twenty-three tumor specimens (NSCLC, breast cancer, neuroblastoma, ovarian cancer, and glioblastoma) from twenty-two patients and were evaluated by a pathologist, placed in media (DMEM, RPMI or RPMI with HEPES) and transported to the laboratory. Briefly, specimens were washed in PBS and then placed in dissociation media. Specimens were dissociated with: collagenase 4 (C4), a mixture of pronase, dispase, elastase and collagenases A and 4 (PDEC), Liberase TM, elastase with DNase I (LE), or Papain. Specimens were minced using spring scissors and underwent two incubation cycles at 37°C for 10 min with rotation followed by further disruption with a pipette. The cells were pelleted and resuspended in ACK lysis buffer, quenched in PBS, and pelleted.  Cells were counted using a Hemocytometer and 8000 cells were loaded into each channel of a 10x Genomics Single-Cell Chromium Controller according to the manufacturer’s instructions.

   Summary of Findings:

   Dissociation of fresh NSCLC specimens with PDEC and LE resulted in the recovery of more fibroblasts and endothelial cells and fewer empty droplets (0.08% and 0.04%, respectively versus 7% with C4). Consistent with cell recovery data, empty droplets were found to cluster in macrophages. However, the dissociation protocol did not affect the number of cells, quality of expression patterns, or CNA patterns in malignant cells.  Overall, C4-dissociated cells had a greater number of detected genes (P = 1.3 × 10⁻⁹⁰ versus PDEC and 1.4 × 10⁻⁶² versus LE), although PDEC-digested cells had a significantly higher number of detectable genes in B-Cells (P = 2 × 10⁻¹⁵ versus C4 and 2 × 10⁻¹⁰ versus LE), and LE-digested cells had the highest number of detectable genes in epithelial cells (P = 5 × 10⁻⁶ versus C4 and 2 × 10⁻⁴ versus PDEC). The authors concluded that dissociation with Papain yielded higher quality malignant cell profiles than CD4 for neuroblastoma specimens. Importantly, the authors state the optimal dissociation protocol is dependent on specimen type.

   Biospecimens

   • Tissue - Breast
   • Tissue - Lung
   • Tissue - Ovary
   • Tissue - Nerve
   • Tissue - Skin
   • Bodily Fluid - Blood

   Preservative Types

   • None (Fresh)

   Diagnoses:

   • Neoplastic - Carcinoma
   • Neoplastic - Melanoma
   • Neoplastic - Pediatric
   • Neoplastic - Sarcoma
   • Neoplastic - Leukemia

   Platform:

   Analyte	Technology Platform
   Cell count/volume	Light microscopy
   RNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Analyte Extraction and Purification	Cell/tissue permeabilization	Dissociated in C4 Dissociated in LE Dissociated in PDEC Dissociated in papain

   View more

2. Study Purpose

   The purpose of this study was to identify potential differences associated with CD-45 depletion on cellular composition by single cell RNA sequencing (scRNA-Seq) of NCSLC, ovarian ascites and blood specimens. A NSCLC tumor specimen was placed in media (DMEM, RPMI or RPMI with HEPES) and transported to the laboratory. The specimen was placed in PDEC dissociation media, minced using spring scissors and underwent two incubation cycles at 37°C for 10 min with rotation followed by further disruption with a pipette. The cells were pelleted and resuspended in ACK lysis buffer, quenched in PBS, and pelleted.  Ovarian ascites were collected from one patient and blood was collected from two patients with CLL. Ascites specimens were centrifuged at 580 g for 5 min at 4°C, the pellet was then resuspended in ACK and lysis buffer and incubated for 5 min before being quenched with PBS, and centrifuged again at 580 g for 5 min at 4°C. The pellet was then resuspended in PBS and filtered using a 70 µm cell strainer. The ACK step was then repeated for ascites if no erythrocytes were present. CLL cells were isolated from peripheral blood mononuclear cells using Ficoll density centrifugation and then snap frozen in liquid nitrogen. CD-45 depletion was performed using CD-45 microbeads and confirmed by flow cytometry. Cells were counted using a Hemocytometer and 8000 cells were loaded into each channel of a 10x Genomics Single-Cell Chromium Controller according to the manufacturer’s instructions.

   Summary of Findings:

   CD-45+ depletion of fresh NSCLC was shown to be important as it increased epithelial cell sequencing of NSCLCs from 26% to 82%.  Depletion of ovarian ascites increased the epithelial cell sequencing from <1% to 32%. Consistent with scRNA-Seq results, Flow cytometry showed an increase in CD-45- cells from 0.75% to 29.4% after CD-45 depletion.

   Biospecimens

   • Tissue - Lung
   • Bodily Fluid - Peritoneal Fluid
   • Cell - Blood

   Preservative Types

   • Frozen
   • OCT

   Diagnoses:

   • Neoplastic - Leukemia
   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   RNA	Next generation sequencing
   Cell count/volume	Flow cytometry
   Cell count/volume	Light microscopy

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Analyte Extraction and Purification	Analyte purification	CD-45 depleted Not depleted

   View more

3. Study Purpose

   The purpose of this study was to identify potential effects of nucleus isolation method on snRNA-Seq of frozen tumor specimens. Neuroblastoma specimens were frozen on dry ice and stored at -80°C. Additional specimens included ovarian tumor and metastatic breast cancer specimens that were embedded in optimal cutting temperature compound (OCT), and snap-frozen sarcoma, glioma, and melanoma specimens from various tissue banks. For specimens in OCT, OCT was reduced by manually removing the periphery and any residual OCT was washed away in PBS. Frozen tissues/cells were then placed in six well plates with Nonidet P40 with salts and Tris (NST), CHAPS, with salts and Tris (CST), Tween with salts and Tris (TST). Tissues were chopped using spring-loaded scissors and cells were pipetted up and down.  The homogenized solutions were strained using a 40 μm Falcon cell strainer and centrifuged at 500 g for 5 min at 4°C. Nuclei isolated using EZlysis were homogenized using a tissue grinder in EZlysis buffer and centrifuged at 500 g for min washed. Pellets from each method were resuspended and filtered using a 35 µm cell strainer. The isolated nuclei were counted using a hemocytometer and sequenced using Chromium Chips for the Chromium Single Cell 3′ Library kit on a HiESeq or NextSeq.

   Summary of Findings:

   Overall nucleus quality and CNA profiles were comparable among nuclei isolated from frozen specimens using TST, CST, or NST although differences were observed among some specimens. Generally, nuclei isolated using EZprep had the lowest number of detectable genes, UMIs, and mitochondrial genes per nuclei. Similarly, TST-isolated nuclei had more nuclei per cell type, and higher levels of mitochondrial genes.  While all methods yielded primarily neuroendocrine cells as expected,  the greatest cell diversity with the largest increase in  recovery of endothelial cells, fibroblasts, neural crest cells and T cells was observed when nuclei were isolated with TST, although high diversity and recovery was also observed with NST isolation. While the authors selected TST as superior for most tumor types, for glioblastomas the authors identified CST as the preferred isolation method (no comparison data shown).

   Biospecimens

   • Tissue - Breast
   • Tissue - Ovary
   • Cell - Blood
   • Tissue - Skin
   • Tissue - Nerve

   Preservative Types

   • Frozen

   Diagnoses:

   • Neoplastic - Carcinoma
   • Neoplastic - Pediatric
   • Neoplastic - Sarcoma
   • Neoplastic - Melanoma
   • Neoplastic - Leukemia

   Platform:

   Analyte	Technology Platform
   Cell count/volume	Light microscopy
   RNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Analyte Extraction and Purification	Analyte isolation method	EZPrep Nonidet P40 with salts and Tris (NST) CHAPS with salts and Tris (CST) Tween with salts and Tris (TST)

   View more

4. Study Purpose

   The purpose of this study was to compare scRNA-Seq and snRNA-Seq data in case-matched tumor specimens. Neuroblastoma and metastatic breast cancer tumor specimens, and blood from a patient with chronic Lymphocytic leukemia were processed as described in studies 1 and 2. Briefly, to generate scRNA- seq data the neuroblastoma specimen was dissociated in Papain, the breast cancer specimens was dissociated in LE, and the CLL specimen was thawed. Cells were pelleted and resuspended in ACK lysis buffer, quenched in PBS, and re-pelleted.   To generate snRNA-Seq data frozen tissues/cells were homogenized in TST, strained using a 40 μm Falcon cell strainer, centrifuged at 500 g for 5 min at 4°C, then resuspended and filtered using a 35 µm cell strainer. Isolated cells and nuclei were counted using a hemocytometer and sequenced using a 10x Genomics Single-Cell Chromium Controller according to the manufacturer’s instructions.

   Summary of Findings:

   Although the same cell types were identified by single cell sequencing and single nuclei sequencing of case-matched neuroblastoma, metastatic breast cancer, and chronic Lymphocytic leukemia specimens, the proportions of each cell type differed between the sequencing methods. However, after batch correction by canonical correlation analysis the cell and nuclei data were mostly aligned. Cells were more likely than nuclei to be in agreement with a previously published dissociation signature (P< 1× 10−100), but for both cells and nuclei the dissociation signature was most prominent in immune and stroma cells.

   Biospecimens

   • Tissue - Breast
   • Cell - Blood
   • Tissue - Nerve

   Preservative Types

   • None (Fresh)
   • Frozen

   Diagnoses:

   • Neoplastic - Carcinoma
   • Neoplastic - Pediatric
   • Neoplastic - Leukemia

   Platform:

   Analyte	Technology Platform
   RNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Preservation	Type of fixation/preservation	Frozen None (fresh)
   Next generation sequencing Specific	Template modification	Isolated cells Isolated nuclei

   View more